sqlsurfing (10/8/2013)
I was curious for the SQLIO test was the test on random and sequential read/writes? Were they all same values or different in SQL IO?
8 threads writing for 120 secs to file D:\TestFile.dat
using 8KB random IOs
throughput metrics:
IOs/sec: 6027.19
MBs/sec: 47.08
latency metrics:
Min_Latency(ms): 0
Avg_Latency(ms): 10
Max_Latency(ms): 263
8 threads reading for 120 secs from file D:\TestFile.dat
using 8KB random IOs
throughput metrics:
IOs/sec: 7349.60
MBs/sec: 57.41
latency metrics:
Min_Latency(ms): 0
Avg_Latency(ms): 8
Max_Latency(ms): 233
8 threads writing for 120 secs to file D:\TestFile.dat
using 64KB sequential IOs
throughput metrics:
IOs/sec: 8265.63
MBs/sec: 516.60
latency metrics:
Min_Latency(ms): 0
Avg_Latency(ms): 7
Max_Latency(ms): 228
8 threads reading for 120 secs from file D:\TestFile.dat
using 64KB sequential IOs
throughput metrics:
IOs/sec: 17319.85
MBs/sec: 1082.49
latency metrics:
Min_Latency(ms): 0
Avg_Latency(ms): 3
Max_Latency(ms): 99
The results from SQLIO are greater than what I calculate the array can provide because I was using 8 threads, I presume?
Steve Jones - SSC Editor (10/7/2013)
You can get a trace from your existing system and run that.
What do I use to get a 'trace' and will it tell me how much tempdb is utilized?
...and thanks for your help with this!